WO2019001521A1 - Data storage method, storage device, client and system - Google Patents

Abstract

The embodiments of the present application relate to the field of computer technologies, and disclosed thereby are a data storage method, a storage device, a client and a system. The method is applied to a first storage device of a storage system, and the first storage device comprises a plurality of solid-state drives; the first storage device receives data sent by a client and performs strip virtualization on the data to obtain a plurality of strip data, the data sent by the client being original data written therein; the first storage device encodes the plurality of strip data by means of an elliptic curve (EC) algorithm to obtain EC strip data, the EC strip data comprising a plurality of strip data and check data, and the check data being used for checking the plurality of strip data; and the first storage device caches a plurality of strip data and check data comprised in the EC strip data into a plurality of solid-state drives respectively. Hence, EC strip data may be recovered by means of the EC algorithm when data are lost, thus ensuring the integrity of data storage.

Description

Data storage method, storage device, client and system

This application claims the priority of the Chinese Patent Application, which is filed on June 29, 2017, the Chinese National Intellectual Property Office, the application number is 201710515890.3, and the invention is entitled "Data Storage Method, Apparatus and System", the entire contents of which are incorporated herein by reference. In the application.

Technical field

The embodiments of the present invention relate to the field of computer technologies, and in particular, to a data storage method, a storage device, a client, and a system.

Background technique

With the rapid development of computer technology, most of the current storage systems are used to implement data storage. The storage system mainly includes a client and a storage device, wherein the client is mainly used for writing data by a user, and the storage device is mainly used for storing data written by a user through a client.

In actual implementation, the storage device generally includes a plurality of solid state disks and a plurality of disks, wherein the solid state disks have high random write and read performance, but the storage capacity is small, and the storage capacity of the disks is large. , but the read and write performance is usually poor. At present, in order to improve the read and write performance of data, in the data storage process, the storage device generally caches the data written by the user delivered by the client to a plurality of solid state disks, and caches any solid disk in the plurality of solid state disks. When the data reaches a certain capacity, the data cached in the plurality of SSDs is stored in the plurality of disks, thereby realizing data storage.

However, when data storage is performed in the above manner, if an SSD is abnormal, the data cached on the SSD will be lost, and the lost data will not be retrieved.

Summary of the invention

The embodiment of the present application provides a data storage method, a storage device, a client, and a system, which can solve the problem that the solid state hard disk abnormality causes data loss and cannot be retrieved. The technical solution is as follows:

The first aspect provides a data storage method, which is applied to a first storage device of a storage system, where the first storage device includes a plurality of solid state disks, and the method includes:

Receiving, by the first storage device, data sent by the client, and performing virtualized strip processing on the data to obtain a plurality of stripe data, where the data sent by the client is the original data written in the client ;

The first storage device performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, The verification data is used to verify the plurality of stripe data;

The first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives, respectively.

The second aspect provides a data storage method, which is applied to a client of a storage system, where the storage system further includes a first storage device, where the first storage device includes a plurality of solid state disks, and the method includes:

The client performs virtualized strip processing on the data to obtain a plurality of stripe data, where the data is original data written in the client;

The client performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, the school The test data is used to verify the plurality of stripe data;

Transmitting, by the client, the EC strip data to the first storage device, so that the first storage device uses the plurality of stripe data and the check data included in the EC strip data Cache to the plurality of solid state drives respectively.

A third aspect provides a data storage method, which is applied to a first storage device of a storage system, where the storage system further includes a client, where the first storage device includes a plurality of solid state disks, and the method includes:

Receiving, by the first storage device, the EC strip data sent by the client, where the EC strip data includes the plurality of stripe data and check data, where the check data is used for the multiple strips Carrying data for verification, the plurality of stripe data being obtained by the client performing virtualized strip processing on the data, where the data is original data written in the client;

The first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives, respectively.

A fourth aspect provides a data storage device, which is deployed in a first storage device of a storage system, where the first storage device includes a plurality of solid state disks, and the device includes:

a receiving processing module, configured to receive data sent by the client, and perform virtualized strip processing on the data to obtain a plurality of stripe data, where the data sent by the client is the original data written in the client ;

An encoding processing module, configured to perform encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and verification data, The verification data is used to verify the plurality of stripe data;

And a first storage module, configured to cache the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives respectively.

In a fifth aspect, a data storage device is provided, which is deployed in a client of a storage system, the storage system further includes a first storage device, the first storage device includes a plurality of solid state disks, and the device includes:

a stripe processing module, configured to perform virtualized stripe processing on the data to obtain a plurality of stripe data, where the data is original data written in the client;

An encoding processing module, configured to perform encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and verification data, The verification data is used to verify the plurality of stripe data;

a sending module, configured to send the EC strip data to the first storage device, so that the first storage device uses the plurality of stripe data included in the EC strip data and the checksum The data is separately cached into the plurality of solid state drives.

The sixth aspect provides a data storage device, which is deployed in a first storage device of the storage system, where the first storage device includes a plurality of solid state disks, and the device includes:

a receiving module, configured to receive EC strip data sent by the client, where the EC strip data includes the plurality of stripe data and check data, and the check data is used to the plurality of strips The data is verified, and the plurality of stripe data is obtained by the client performing virtualized strip processing on the data, where the data is original data written in the client;

And a first storage module, configured to cache the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives respectively.

The seventh aspect provides a first storage device, which is deployed in a storage system, where the first storage device includes a plurality of solid state disks, and the first storage device includes:

One or more processors; and

Memory

The memory stores one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including instructions for:

Receiving data sent by the client, and performing virtualized strip processing on the data to obtain a plurality of stripe data, where the data sent by the client is the original data written in the client;

Encoding the plurality of stripe data by using a codec EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, and the check data is used for Performing verification on the plurality of stripe data;

And storing the plurality of stripe data and the check data included in the EC strip data into the plurality of solid state drives respectively.

The eighth aspect provides a client that is deployed in a storage system, where the storage system further includes a first storage device, where the first storage device includes a plurality of solid state disks, and the client includes:

One or more processors; and

Memory

The memory stores one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including instructions for:

Performing virtualized stripe processing on the data to obtain a plurality of stripe data, the data being original data written in the client;

Encoding the plurality of stripe data by using a codec EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, and the check data is used for Performing verification on the plurality of stripe data;

Transmitting the EC stripe data to the first storage device, so that the first storage device caches the plurality of stripe data and the check data included in the EC stripe data respectively Said in multiple solid state drives.

A ninth aspect, a first storage device is provided, which is deployed in a storage system, where the first storage device includes a plurality of solid state disks, and the first storage device includes:

One or more processors; and

Memory

The memory stores one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including instructions for:

Receiving EC strip data sent by the client, the EC strip data includes the plurality of stripe data and check data, and the check data is used to verify the plurality of stripe data, The plurality of stripe data is obtained by the client performing virtualized strip processing on the data, where the data is original data written in the client;

And storing the plurality of stripe data and the check data included in the EC strip data into the plurality of solid state drives respectively.

According to a tenth aspect, a storage system is provided, where the storage system includes a client and a first storage device, and the first storage device includes a plurality of solid state disks, and the method includes:

The client performs virtualized strip processing on the data to obtain a plurality of stripe data, where the data is original data written in the client;

The client performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, the school The test data is used to verify the plurality of stripe data;

Sending, by the client, the EC strip data to the first storage device;

Receiving, by the first storage device, the EC strip data sent by the client, and buffering the plurality of strip data and the check data included in the EC strip data to the plurality of solid state drives respectively in.

The beneficial effects brought by the technical solutions provided by the embodiments of the present application are:

In the storage system, when the first storage device receives the data written by the client and is written in the client, the data is sequentially subjected to virtualized strip processing and encoding processing, thereby obtaining a plurality of stripe data and Check the EC strip data of the data. Then, the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives. In this way, when the data is lost, the EC stripe data can be recovered by the EC algorithm, thereby ensuring the integrity of the data storage.

DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings used in the description of the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the present application. Other drawings may also be obtained from those of ordinary skill in the art in light of the inventive work.

FIG. 1 is a schematic structural diagram of a storage system according to an exemplary embodiment.

FIG. 2 is a flow chart showing a data storage method according to an exemplary embodiment.

FIG. 3 is a schematic diagram of a specific implementation of an EC algorithm according to the embodiment of FIG. 2. FIG.

4 is a schematic diagram of a specific implementation of another EC algorithm involved in the embodiment of FIG. 2.

FIG. 5 is a schematic diagram of a specific implementation of another EC algorithm involved in the embodiment of FIG. 2. FIG.

FIG. 6 is a schematic diagram of a specific implementation of another EC algorithm involved in the embodiment of FIG. 2. FIG.

FIG. 7 is a schematic diagram of a data storage method according to the embodiment of FIG. 2. FIG.

FIG. 8 is a flowchart of a data storage method according to another exemplary embodiment.

FIG. 9 is a schematic structural diagram of a data storage device according to an exemplary embodiment.

FIG. 10 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 11 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 12 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 13 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 14 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 15 is a schematic structural diagram of a data storage device according to an exemplary embodiment.

FIG. 16 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 17 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 18 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 19 is a schematic structural diagram of a data storage device according to another exemplary embodiment.

FIG. 20 is a schematic structural diagram of a first storage device according to an exemplary embodiment.

FIG. 21 is a schematic structural diagram of a client according to an exemplary embodiment.

Detailed ways

The embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

Before the detailed description of the embodiments of the present application, the nouns involved in the embodiments of the present application are briefly introduced:

The EC algorithm: that is, the (Erasure Code) algorithm is a data protection method, which can be mainly used for performing redundant data block expansion and encoding on data.

Solid state drives: Solid State Drives (SSDs), which are typically fabricated from solid state electronic memory chip arrays, consisting primarily of control units and memory cells. The storage unit includes, but is not limited to, a FLASH chip and a DRAM (Dynamic Random Access Memory) chip. Solid state drives have high random write and read and write performance, but the capacity is generally small and expensive. Typically, this solid state drive is used for higher levels of tiered storage, ie as a cache device.

Disk: It can read and write data through mechanical operations such as disk rotation and head movement, and has a large storage capacity. However, due to mechanical operation such as disk rotation and head movement, disk read and write performance is poor. . Disks are typically used for lower layers of hierarchical storage, ie as the final carrier for data storage. In general, you can use the copy method for data storage.

Hierarchical storage: refers to the hierarchical storage of data based on factors such as read and write performance, business continuity, storage security, data retention policies, and hardware costs. For example, hierarchical storage methods may include primary storage, backup disk storage, and archive disk. Storage, disk archive storage, CD archive storage, and more. That is to say, the tiered storage is based on the importance of the data, the access frequency, the retention time, the capacity, the performance, and the like, and the data is stored in different storage modes on the first storage device of different performance. The grading storage can realize automatic migration of data between the SSD and the disk. Specifically, the first storage device can implement data migration according to the frequency of data being accessed, for example, migrating data with less frequent access frequency from the SSD. To disk, or to migrate data that is accessed more frequently (which can be called hotspot data) from disk to SSD.

Next, the storage system according to the embodiment of the present application is introduced. Please refer to FIG. 1 , which is a schematic structural diagram of a storage system. The storage system mainly includes a client 110 and a plurality of storage devices, and the plurality of storage devices include a first Storage device 120. The client 110 can be connected to the first storage device 120 through a wired network or a wireless network.

The client 110 is mainly used for user to write data, and the client 110 can support keyboard input. The keyboard can be a virtual keyboard or a physical keyboard. In an actual application scenario, the client 110 can be installed in the terminal and run the client 110 through the terminal. The terminal may be a device such as a mobile phone, a tablet computer, or a computer, which is not limited in this embodiment of the present application.

It should be noted that the client 110 may not be installed in the terminal, but may be a physical device.

The first storage device 120 is mainly configured to receive data sent by the client 110 and store the data. The first storage device 120 includes a plurality of solid state disks. In a specific implementation, the first storage device 120 can implement data caching through the plurality of solid state disks.

In addition, in a possible implementation, the first storage device 120 may further include multiple disks, and in some conditions, the first storage device 120 stores the data cached in the plurality of solid state disks to the plurality of disks. In the disk, the specific implementation is as described in the following FIG. 2 or FIG. 8 embodiment.

In another possible implementation manner, the multiple storage devices further include a second storage device 130, where the second storage device 130 includes a plurality of disks. That is, in an actual implementation, the plurality of SSDs and the plurality of disks may not be in one storage device. For example, the plurality of SSDs belong to the first storage device 120, and the plurality of disks belong to the second storage device 130. . In this case, the process implementation process for data involved below is performed by the first storage device 120, and then, under some conditions, the first storage device 120 transmits data stored by the plurality of solid state disks to the second The storage device 130 stores the data by the second storage device 130 through a plurality of disks included therein.

In an actual implementation, the embodiment of the present application first performs virtualized strip processing on data written in the client 110 and performs encoding processing by using an EC algorithm to obtain EC strip data, and then EC stripe data is cached into multiple SSDs. When the data cached in the plurality of SSDs meets certain conditions, the data cached in the plurality of SSDs is transferred to the disk for storage. In this way, since the solid state hard disk is used as the intermediate cache, the data read and write performance is improved, and since the encoding process is performed by the EC algorithm, and the principle of the EC algorithm is to add check data to the data, when data loss occurs In this case, based on the EC algorithm, the verification data can be used for data recovery to ensure the reliability of data storage. Generally, the manner of tiered storage is referred to as distributed storage. For the specific implementation, please refer to the embodiment of FIG. 2 or FIG. 8 below.

Further, in a specific implementation, the storage system may further include an MDS (Meta Data Server) 140, and the metadata server 140 is mainly used for allocating and managing storage of data. In effect, the metadata server 140 is equivalent to the dispatch center of the storage system, i.e., the metadata server 140 can control which solid storage disk or disk to store the data to. In a specific implementation, the metadata server 140 can monitor the storage state of the first storage device 120, and allocate the solid state hard disk to the client 110 according to the monitored storage state. For the specific implementation process, refer to the embodiments described below.

It should be noted that the embodiment of the present application is only described by using the metadata server 140 in the storage system as an example. In another embodiment, if the metadata server 140 is not included in the storage system, the storage of the data may be controlled by specifying a scheduling algorithm, where the specified scheduling algorithm may be run on the client or in the first On the storage device, this embodiment of the present application does not introduce too much.

As described above, in the embodiment of the present application, a series of processes such as virtualized striping and encoding are required for data written on the client. In an actual implementation, this series of processing may be performed by the first storage device or by the client. Next, the data storage method will be described in detail by taking the first storage device and the client to perform the series of processing as an example, as shown in the following FIG. 2 embodiment and FIG.

Please refer to FIG. 2 , where the above series of processing is performed by the first storage device as an example for description. FIG. 2 is a flowchart of a data storage method according to an exemplary embodiment. The embodiment may be implemented by using the storage system shown in FIG. 1. The data storage method may include the following steps:

Step 201: The client sends a storage request to the metadata server.

In an actual implementation, when the client needs to store data to the first storage device, a storage request may be sent to the metadata server, where the storage request is used to instruct the metadata server to allocate a solid state hard disk for storing data for the client. Further, in order to facilitate the metadata server to determine that several SSDs need to be allocated to the client, the storage request may generally carry the number of copies of the data to be stored, so that the metadata server can determine the number of copies according to the number of copies. The number of SSDs that need to be allocated.

Step 202: The metadata server allocates a solid state hard disk to the client.

Generally, since the metadata server can obtain the storage space size of the first storage device, and the metadata server is responsible for allocating a storage location of the data to the client, that is, the metadata server can monitor the first storage device. The storage server can obtain the storage status of the first storage device. Further, the metadata server can obtain the current storage status of the plurality of SSDs. Therefore, the metadata server may allocate a solid state hard disk to the client according to the number of copies carried in the storage request and the current storage condition of each solid state hard disk in the plurality of solid state disks, and send the address information of the allocated solid state disk. Give the client.

In a specific implementation, the metadata server may send a storage response to the client, where the storage response carries address information of the allocated SSD.

Step 203: The client sends data to the first storage device, where the first storage device receives data sent by the client, and the data sent by the client is data written in the client.

As described above, in this embodiment, the data written in the client is processed through the first storage device, so that the client can directly send the write to the client to the first storage device. The original data, in a specific implementation, the client may send data to the first storage device based on the address information of the solid state hard disk allocated by the metadata server.

Step 204: The first storage device performs virtualized strip processing on the data to obtain a plurality of stripe data.

After receiving the data sent by the client, the first storage device performs virtualized stripe data processing on the data. In a specific implementation, the first storage device divides the data into the same according to a preset fixed stripe block size. Multiple pieces of data of a size, the multiple pieces of data of the same size obtained here are referred to as a plurality of stripe data. For example, if the data is 5 M data, the first storage device divides the data according to the 1 M strip block size, and after the virtualized strip processing, the data is divided into 5 pieces to obtain 5 strip data. The size of each stripe data is 1M. For example, the five strip data are D1, D2, D3, D4, and D5, respectively.

Step 205: The first storage device performs encoding processing on the plurality of stripe data by using an EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, and the check data is used by the check data. The plurality of stripe data is verified.

In a specific implementation, the first storage device generates a vector based on the plurality of strip data, and multiplies the generated vector by a preset generation matrix to obtain a coding vector, where the first storage device determines the data included in the coding vector. Strip data for this EC.

In the actual algorithm operation, the specific implementation process of the first storage device to generate a vector based on the multiple stripe data includes: the first storage device constructs the plurality of stripe data into a column vector.

The preset generation matrix can be customized by the user according to actual needs.

For example, referring to FIG. 3, the preset generation matrix is a matrix B. If the plurality of stripe data are D1, D2, D3, D4, and D5, a vector D is generated based on the plurality of stripe data, and the generated vector D is multiplied by the preset generation matrix B to obtain a coded vector C. The first storage device determines the data included in the code vector C as EC stripe data. Specifically, the plurality of stripe data included in the EC stripe data are D1, D2, D3, D4, and D5, respectively, and the check data is C1, C2, and C3.

It should be noted that, according to the setting rules of the preset generation matrix, the number of the verification data is also different. In an actual implementation, the technical personnel can perform the setting according to the actual requirements, which is not limited by the embodiment of the present application.

Further, when a data loss situation occurs in the plurality of solid state disks, determining a target row from the preset generation matrix, wherein the preset data in the target row is multiplied by a vector composed of the plurality of stripe data The lost data can be obtained. After the first storage device deletes the target row in the preset generation matrix, the target generation matrix is obtained, and the vector composed of the EC strip data remaining after the data is lost and the target generation matrix are obtained. The inverse matrix is multiplied to recover the EC strip data.

In the actual application scenario, the data of the cached data of the SSD may be lost due to the poor connection of the plug-in interface of the SSD and the power failure of the first storage device. In the embodiment of the present application, The lost data is recovered by the EC algorithm.

For example, if the three data of D1, D3, and C1 included in the EC strip data are lost, the implementation process of recovering the stripe data included in the EC stripe data is as follows:

Referring to FIG. 4, the target row is determined from the preset generation matrix B. Since the lost data is D1, D3, and C1, the rows in the preset generation matrix B that are multiplied by the vector D composed of the plurality of stripe data to obtain the lost data are the first row and the third row, respectively. And the sixth line, therefore, it can be determined that the target line is the first line, the third line, and the sixth line, respectively. The first row, the third row, and the sixth row in the preset generation matrix B are deleted, and the target generation matrix is obtained as shown in FIG. 4. The first storage device determines an inverse matrix of the target generation matrix according to the formula shown in FIG.

After that, referring to the formula shown in FIG. 5, the first storage device multiplies the vector composed of the EC strip data remaining after the data is lost, and obtains the formula shown in FIG. 6. After the first storage device solves the formula, the lost strip data D1 and the strip data D3 can be determined, thereby realizing recovery of the EC strip data.

It can be seen that after the above series of processing, even if the SSD abnormality causes data loss, the first storage device can recover the lost data through the EC algorithm, thereby ensuring the reliability of the data storage.

Step 206: The first storage device caches the plurality of stripe data and the check data included in the EC stripe data into the plurality of solid state drives.

Referring to FIG. 4, it is assumed that the plurality of stripe data included in the EC stripe data are D1, D2, D3, D4, and D5, and the check data is C1, C2, and C3, and the first storage device uses the EC strip. With data cache to multiple solid state drives as shown in Figure 7.

In an actual implementation, after the plurality of stripe data and the check data included in the EC strip data are respectively cached in the plurality of solid state disks, the first storage success message may be sent to the metadata server to notify The metadata server successfully implements data storage. In addition, in order to facilitate the data management of the metadata server, the first storage success message may further carry the first storage information of the stored data, where the first storage information may include information such as specific storage locations of the data on each solid state drive.

Since the cached data is encrypted in the SSD, the EC stripe data of the check data is added, and the data is not backed up, so the data storage is performed in the SSD compared to the backup mode. Can save storage space for SSDs.

It should be noted that, in an actual implementation, after the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives, the first storage device may, under some conditions, the first storage device Data from multiple SSD caches is stored on disk. The following two possible implementations can be included in the implementation:

The first case: under the preset condition, the first storage device performs data storage by using a plurality of disks included in the first storage device.

In this case, the first storage device further includes a plurality of disks. In a specific implementation, the specific implementation of the data storage by the first storage device by using multiple disks included in the first storage device may include the following steps (1)-(2):

(1) if the data capacity of any of the plurality of solid state hard disks is up to a preset capacity, or if the data of any of the plurality of solid state disks is read in the preset time period If the number of times is less than the preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to the data attribute information of the plurality of stripe data to obtain replica data.

The preset capacity may be customized by the user according to actual requirements, or may be set by default by the first storage device, which is not limited by the embodiment of the present application. For example, the preset capacity may be one-third of the total storage capacity of the solid state drive.

In a specific implementation, the current storage capacity of the plurality of solid state drives may be determined by a metadata server. That is, after the metadata server receives the first storage success message, the current storage capacity of each solid state hard disk can be determined. If the metadata server determines that the data capacity of any of the plurality of SSDs in the plurality of SSDs reaches a preset capacity, the data stored in the plurality of SSDs needs to be moved to the disk.

The preset time period may be customized by the user according to actual needs, or may be set by default by the first storage device, which is not limited by the embodiment of the present application.

The preset number of times may be customized by the user according to actual needs, or may be set by default by the first storage device, which is not limited by the embodiment of the present application.

In a specific implementation, the number of times the data in the plurality of solid state drives is read may be monitored by a metadata server. That is, after the metadata server receives the first storage success message, the number of times the data cached in the plurality of solid state disks is read may be monitored. If the number of times the data of any one of the plurality of SSDs is read less than the preset number of times in the preset time period, the data is cold data, and the storage space of the solid state hard disk may be saved. Data stored in multiple SSDs is moved to disk. Among them, the cold data refers to data that is not often used.

In a possible implementation, the metadata server selects multiple disks according to the current storage state of the multiple disks, and sends address information of the multiple disks to the first storage device, so that the first The storage device stores data in the plurality of solid state drives through the plurality of disks.

After receiving the plurality of address information, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to the data attribute information of the plurality of stripe data to obtain replica data. The data attribute information may be used to indicate information such as a storage attribute, a folder, and the like of the plurality of stripe data.

(2) The first storage device stores the duplicate data in a plurality of disks included in the first storage device.

After the first storage device generates the replica data, the replica data is stored in the plurality of disks. For example, the first storage device generates N replica data, and stores the N replica data into N disks respectively. The N copies of the data are stored as cold data, wherein the N is a natural number greater than one.

Since the data stored in the plurality of disks is stored in the form of a copy, parallel reading can be supported in the process of reading data, thereby improving the performance of reading.

The second case: under the preset condition, the first storage device performs data storage by using a plurality of disks included in the second storage device.

In this case, the first storage device includes only a plurality of solid state disks, and the storage system further includes a second storage device, where the second storage device includes a plurality of disks. In a specific implementation, the specific implementation of the data storage by the first storage device by using multiple disks included in the second storage device may include the following steps (3)-(5):

(3) If the data capacity of any of the plurality of solid state drives is up to a preset capacity, or if the data of any of the plurality of SSDs is read in the preset time period If the number of times is less than the preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to the data attribute information of the plurality of stripe data to obtain replica data.

For the specific implementation, reference may be made to the step (1) in the first case mentioned above, and the detailed description is not repeated here.

(4) The first storage device sends the copy data to the second storage device.

Because the copy data needs to be stored in the disk, the first storage device does not include the disk. Therefore, the first storage device can store the copy data through another second storage device including multiple disks, that is, the first storage device. The copy data is sent to the second storage device.

(5) The second storage device receives the copy data sent by the first storage device, and stores the received copy data into a plurality of disks included in the second storage device.

In a specific implementation, the specific implementation that the second storage device stores the received copy data in the multiple disks included in the second storage device may be referred to that the first storage device stores the copy data to the first storage device. The implementation process in multiple disks.

Step 207: The first storage device deletes the plurality of stripe data and the check data buffered in the plurality of solid state drives.

In an actual implementation, in order to save the storage space of the solid state hard disk, after the plurality of stripe data is successfully stored in the disk, the plurality of stripe data cached in the plurality of solid state disks and the check data may be deleted. . In a specific implementation, after the first storage device successfully stores the replica data through the plurality of disks included in the first storage device, or when the first storage device successfully stores the replica data through multiple disks included in the second storage device Afterwards, the first storage device deletes the plurality of stripe data and the check data buffered in the plurality of solid state drives.

After the first storage device successfully stores the replica data through the plurality of disks included in the first storage device, the first storage device can successfully store the replica data in the plurality of disks included in the first storage device. The data storage server sends a second storage success message, where the second storage success message is used to notify the metadata server that the copy data is successfully stored in the disk. Further, the second storage success message may carry the second storage information. The second storage information may include information such as a specific storage location of the duplicate data on the disk.

After receiving the second storage success message, the metadata server may return a deletion indication to the first storage device, where the deletion indication may be used to instruct the first storage device to delete data cached in the plurality of solid state disks. That is, after receiving the deletion indication, the first storage device deletes the plurality of stripe data and the plurality of verification data buffered in the plurality of solid state disks.

Alternatively, after the first storage device successfully stores the replica data through the plurality of disks included in the second storage device, after the second storage device successfully stores the replica data in the plurality of disks included in the second storage device, Sending a third storage success message to the metadata server, the third storage success message is used to notify the metadata server that the copy data is successfully stored in the disk, and further, the third storage success message may carry the third storage information. The third storage information may include information such as a specific storage location of the replica data on the disk.

After receiving the third storage success message, the metadata server may return a deletion indication to the first storage device, where the deletion indication may be used to instruct the first storage device to delete data cached in the plurality of solid state disks. That is, after receiving the deletion indication, the first storage device deletes the plurality of stripe data and the plurality of verification data buffered in the plurality of solid state disks.

It should be noted that, in the process of reading data, there is a case that the first storage device stores the replica data generated based on the plurality of stripe data on the disk, and the cached in the plurality of solid state disks The data has not yet been deleted. Therefore, in order to improve the performance of reading data, the client can generally read data from the solid state hard disk first, and if the data cached in the solid state hard disk is deleted, the data is read from the disk.

In a specific implementation, when the client side needs to read data, the read request may be sent to the metadata server, where the read request carries related information of the data to be read. The metadata server can sense the storage status of each SSD and each disk in the first storage device. After receiving the read request, the metadata server can determine the plurality of SSDs according to the related information. Whether to store the data that needs to be read.

If it is determined that the data to be read is stored in the plurality of solid state disks, obtaining specific storage location information of the data to be read in the plurality of solid state disks, and transmitting the obtained specific storage location information to the client End, so that the client reads data from the plurality of solid state disks according to the specific storage location information.

Otherwise, if the plurality of SSDs do not store data that needs to be read, the metadata server determines, according to the related information, storage location information of the data to be read on the disk, and determines the storage location information. The client is sent so that the client reads data from the plurality of disks based on the storage location information.

In addition, it should be noted that, as described above, in the process of hierarchical storage, the first storage device may migrate some hotspot data from the disk to the solid state hard disk according to the frequency of data access. Among them, the hotspot data refers to data that is often used.

Before the hotspot data is migrated from the disk to the solid state hard disk, the first storage device may perform the above series of processing on the hotspot data by using the operations performed in steps 204 to 205. In this way, after the hotspot data is stored in the SSD, even if the SSD is abnormal, the hotspot data is lost, and the first storage device can retrieve the lost hotspot data, thereby preventing the data storage system from being abnormal due to the loss of the hotspot data. The situation of operation.

In the embodiment of the present application, in the storage system, when the first storage device receives the data written by the client and is written in the client, the data is sequentially subjected to virtualized strip processing and encoding processing, and then obtained. EC stripe data including multiple stripe data and check data. Then, the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives. In this way, when the data is lost, the EC stripe data can be recovered by the EC algorithm, thereby ensuring the integrity of the data storage.

Please refer to FIG. 8 , and then the above series of processing is performed by the client as an example. FIG. 8 is a flowchart of a data storage method according to an exemplary embodiment. The embodiment may be implemented by using the storage system shown in FIG. 1. The data storage method may include the following steps:

Step 301: The client sends a storage request to the metadata server.

For the specific implementation process, refer to step 201 in the foregoing embodiment of FIG. 2.

Step 302: The metadata server allocates a solid state hard disk to the client.

For the specific implementation process, refer to step 202 in the foregoing embodiment of FIG. 2.

Step 303: The client performs virtualized strip processing on the data to obtain a plurality of stripe data.

It should be noted that, for the specific implementation of the virtualized stripe processing of the data by the client, refer to the specific implementation process of the virtualized stripe processing of the data by the first storage device in step 204 in the foregoing embodiment of the present invention. The details are not repeated here.

Wherein, based on the implementation process of the virtualized stripe processing, the virtualized stripe processing may be considered to divide the continuous data into data blocks of the same size.

Step 304: The client performs encoding processing on the plurality of stripe data by deleting the correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, and the check data Used to verify the multiple stripe data.

That is, the client generates a vector based on the plurality of stripe data, and multiplies the generated vector by a preset generation matrix to obtain a code vector, and the client determines the data included in the code vector as the EC stripe. data.

For a specific implementation process, reference may be made to the specific implementation of the encoding processing of the plurality of stripe data by the first storage device in the step 205 in the embodiment of FIG. 2, which is not described in detail herein.

Further, when a data loss situation occurs in the plurality of solid state disks, the first storage device determines a target row from the preset generation matrix, and the preset data in the target row and the vector of the plurality of stripe data After multiplication, you can get the missing data. The first storage device deletes the target row in the preset generation matrix to obtain a target generation matrix, and multiplies a vector composed of EC strip data remaining after data loss by an inverse matrix of the target generation matrix to EC strip data is restored.

For a specific implementation process of recovering the EC strip data, refer to the specific description of the foregoing embodiment of FIG. 2, and details are not described herein again.

It should be noted that the specific implementation principle that the client performs the above-mentioned series of processing on the written data is the same as the specific implementation principle in which the first storage device performs a series of processing on the data sent by the client. Therefore, it will not be described in detail here.

Step 305: The client sends the EC stripe data to the first storage device, so that the first storage device caches the plurality of stripe data and the check data included in the EC stripe data to the plurality of Solid state drive.

After the data is processed by the client, the EC strip data obtained by the processing may be sent to the first storage device. In a specific implementation, the client may use the address information of the solid state hard disk allocated by the metadata server. The first storage device sends the EC strip data obtained after the processing.

Step 306: The first storage device receives the EC strip data sent by the client, where the EC strip data includes the plurality of strip data and check data, where the check data is used to perform the multiple strip data. Verification, the plurality of stripe data is obtained by the client performing virtualized strip processing on the data, and the data is the original data written in the client.

Step 307: The first storage device caches the plurality of stripe data and the check data included in the EC stripe data into the plurality of solid state drives.

For the specific implementation process, refer to step 206 in the foregoing embodiment of FIG. 2.

In addition, in the actual implementation, after the first storage device caches the plurality of stripe data and the check data included in the EC stripe data into the plurality of solid state hard disks, the first storage may be performed under some conditions. The device stores the data of the plurality of SSD caches on the disk. The following two possible implementations can be included in the specific implementation:

The first case: under the preset condition, the first storage device performs data storage by using a plurality of disks included in the first storage device.

In this case, the first storage device further includes a plurality of disks. In a specific implementation, the specific implementation of the data storage by the first storage device by using multiple disks included in the first storage device may include the following steps (6)-(7):

(6) if the data capacity of any of the plurality of solid state drives is up to a preset capacity, or if the data of any of the plurality of SSDs is read during the preset time period If the number of times is less than the preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to the data attribute information of the plurality of stripe data to obtain replica data.

For the specific implementation process, refer to step (1) in the foregoing embodiment of FIG. 2.

(7) The first storage device stores the copy data in a plurality of disks included in the first storage device.

For the specific implementation process, refer to step (2) in the foregoing embodiment of FIG. 2.

The second case: under the preset condition, the first storage device performs data storage through a plurality of disks included in the second storage device.

In this case, the first storage device includes only a plurality of solid state disks, and the storage system further includes a second storage device, where the second storage device includes a plurality of disks. In a specific implementation, the specific implementation of the data storage by the first storage device by using multiple disks included in the second storage device may include the following steps (8)-(10):

(8) If the data capacity of any of the plurality of solid state drives is up to a preset capacity, or if the data of any one of the plurality of SSDs is read in the preset time period If the number of times is less than the preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to the data attribute information of the plurality of stripe data to obtain replica data.

For the specific implementation process, refer to step (3) in the foregoing embodiment of FIG. 2.

(9) The first storage device sends the copy data to the second storage device.

For the specific implementation process, refer to step (4) in the foregoing embodiment of FIG. 2.

(10) The second storage device receives the copy data sent by the first storage device, and stores the received copy data into a plurality of disks included in the second storage device.

For the specific implementation process, refer to step (5) in the foregoing embodiment of FIG. 2.

Step 308: The first storage device deletes the plurality of stripe data and the check data buffered in the plurality of solid state drives.

For the specific implementation process, refer to step 207 in the foregoing embodiment of FIG. 2.

In the embodiment of the present application, in the storage system, after the virtualized stripe processing and the encoding process are sequentially performed on the data written by the client, the EC strip data including the plurality of stripe data and the check data is obtained. Then, the EC strip data is sent to the first storage device, and the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives. In this way, when the data is lost, the EC stripe data can be recovered by the EC algorithm, thereby ensuring the integrity of the data storage.

FIG. 9 is a schematic structural diagram of a data storage device, which may be implemented by software, hardware, or a combination of both, according to an exemplary embodiment. The data storage device is deployed in a first storage device in the storage system shown in FIG. 1 , where the first storage device includes a plurality of solid state disks, and the device includes:

The receiving processing module 410 is configured to receive data sent by the client, perform virtualized strip processing on the data, and obtain a plurality of stripe data, where the data sent by the client is written in the client Raw data;

The encoding processing module 412 is configured to perform encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and verification data, The verification data is used to verify the plurality of stripe data;

The first storage module 414 is configured to cache the plurality of stripe data and the verification data included in the EC stripe data into the plurality of solid state drives.

Optionally, the first storage device further includes multiple disks. Referring to FIG. 10, the device further includes:

The first aggregation module 415 is configured to: if the data capacity of any one of the plurality of solid state disks is up to a preset capacity, or if any of the plurality of SSDs is cached within a preset time period The number of times the data is read is less than the preset number of times, and the plurality of stripe data included in the EC stripe data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data;

The second storage module 416 stores the copy data into a plurality of disks included in the first storage device.

Optionally, referring to FIG. 11, the apparatus further includes:

The first deleting module 418 is configured to delete the plurality of stripe data and the check data buffered in the plurality of solid state disks.

Optionally, the encoding processing module 412 is configured to:

Generating a vector based on the plurality of stripe data, and multiplying the generated vector by a preset generation matrix to obtain a coding vector;

The data included in the coding vector is determined as the EC strip data.

Optionally, referring to FIG. 12, the apparatus further includes:

a determining module 420, configured to determine, from the preset generation matrix, a target row, a preset data in the target row, and a vector consisting of the multiple stripe data when a data loss situation occurs in the plurality of solid state drives After multiplication, the lost data can be obtained;

a second deleting module 421, configured to delete the target row in the preset generation matrix to obtain a target generation matrix;

The recovery module 422 is configured to multiply a vector consisting of the remaining EC strip data after data loss and an inverse matrix of the target generation matrix to recover the EC strip data.

Optionally, the storage system further includes a second storage device, and the second storage device includes a plurality of disks. Referring to FIG. 13, the device further includes:

The second aggregation module 424 is configured to: if the data capacity of any one of the plurality of solid state disks is up to a preset capacity, or if any of the plurality of SSDs is cached within a preset time period The number of times the data is read is less than the preset number of times, and the plurality of stripe data included in the EC stripe data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data;

a sending module 426, configured to send the replica data to the second storage device; the second storage device receives replica data sent by the sending module, and stores the received replica data to the second storage The device is included in multiple disks.

In the embodiment of the present application, in the storage system, when the first storage device receives the data written by the client and is written in the client, the data is sequentially subjected to virtualized strip processing and encoding processing, and then obtained. EC stripe data including multiple stripe data and check data. Then, the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives. In this way, when the data is lost, the EC stripe data can be recovered by the EC algorithm, thereby ensuring the integrity of the data storage.

FIG. 14 is a schematic structural diagram of a data storage device according to an exemplary embodiment. The data storage device can be implemented by software, hardware, or a combination of both. The data storage device is deployed in a client in the storage system shown in FIG. 1. The storage system further includes a first storage device, the first storage device includes a plurality of solid state disks, and the device includes:

The stripe processing module 510 is configured to perform virtualized stripe processing on the data to obtain a plurality of stripe data, where the data is original data written in the client;

The encoding processing module 530 is configured to perform encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data. The verification data is used to verify the plurality of stripe data;

The sending module 540 is configured to send the EC strip data to the first storage device, so that the first storage device uses the plurality of strip data and the school included in the EC strip data. The test data is separately cached into the plurality of solid state drives.

Optionally, the encoding processing module 530 is configured to:

Generating a vector based on the plurality of stripe data, and multiplying the generated vector by a preset generation matrix to obtain a coding vector;

The data included in the coding vector is determined as the EC strip data.

In the embodiment of the present application, in the storage system, after the virtualized stripe processing and the encoding process are sequentially performed on the data written by the client, the EC strip data including the plurality of stripe data and the check data is obtained. Then, the EC strip data is sent to the first storage device, and the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives. In this way, when the data is lost, the EC stripe data can be recovered by the EC algorithm, thereby ensuring the integrity of the data storage.

FIG. 15 is a schematic structural diagram of a data storage device according to an exemplary embodiment. The data storage device can be implemented by software, hardware, or a combination of both. The data storage device is deployed in the first storage device in the storage system shown in FIG. 1. The storage system further includes a client, where the first storage device includes a plurality of solid state disks, and the device includes:

The receiving module 610 is configured to receive EC strip data sent by the client, where the EC strip data includes the plurality of strip data and check data, where the check data is used to the multiple strips Carrying data for verification, the plurality of stripe data being obtained by the client performing virtualized strip processing on the data, where the data is original data written in the client;

The first storage module 620 is configured to cache the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives.

Optionally, the first storage device further includes a plurality of disks. Referring to FIG. 16, the device further includes:

The first aggregation module 622 is configured to: if the data capacity of any one of the plurality of solid state disks is up to a preset capacity, or if any of the plurality of SSDs is cached within a preset time period The number of times the data is read is less than the preset number of times, and the plurality of stripe data included in the EC stripe data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data;

The second storage module 624 is configured to store the replica data into a plurality of disks included in the first storage device.

Optionally, referring to FIG. 17, the apparatus further includes:

The first deleting module 630 is configured to delete the plurality of stripe data and the check data buffered in the plurality of solid state disks.

Optionally, referring to FIG. 18, the apparatus further includes:

a determining module 640, configured to determine a target row from the preset generation matrix when a data loss situation occurs in the plurality of solid state disks, where preset data in the target row is combined with the plurality of stripe data The vector is multiplied to get the missing data;

a second deleting module 642, configured to delete the target row in the preset generation matrix to obtain a target generation matrix;

The recovery module 650 is configured to multiply a vector composed of EC strip data remaining after data loss and an inverse matrix of the target generation matrix to recover the EC strip data.

Optionally, the storage system further includes a second storage device, and the second storage device includes a plurality of disks. Referring to FIG. 19, the device further includes:

The second aggregation module 660 is configured to: if the data capacity of any one of the plurality of solid state disks is up to a preset capacity, or if any of the plurality of SSDs is cached within a preset time period The number of times the data is read is less than the preset number of times, and the plurality of stripe data included in the EC stripe data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data;

a sending module 670, configured to send the copy data to the second storage device; the second storage device receives the copy data sent by the first storage device, and stores the received copy data to the first Two storage devices are included in multiple disks.

In the embodiment of the present application, in the storage system, after the virtualized stripe processing and the encoding process are sequentially performed on the data written by the client, the EC strip data including the plurality of stripe data and the check data is obtained. Then, the EC strip data is sent to the first storage device, and the first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives. In this way, when the data is lost, the EC stripe data can be recovered by the EC algorithm, thereby ensuring the integrity of the data storage.

It should be noted that, when the data storage device provided by the foregoing embodiment implements data storage, only the division of each functional module described above is illustrated. In an actual application, the function distribution may be completed by different functional modules as needed. The internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the data storage device and the data storage method embodiment provided by the foregoing embodiments are in the same concept, and the specific implementation process is described in detail in the method embodiment, and details are not described herein again.

FIG. 20 is a schematic structural diagram of a first storage device, which mainly includes a transmitter 701, a receiver 702, a memory 704, a processor 703, and a communication bus 705, according to an exemplary embodiment. It will be understood by those skilled in the art that the structure of the first storage device 700 shown in FIG. 20 does not constitute a limitation on the intelligent lighting device, may include more or less components than those illustrated, or combine some components, or The components of the present application are not limited in this embodiment.

The transmitter 701 can be configured to send data, such as frame data, to the subsequent processing module.

The memory 704 can be used to store parsed valid data, and the memory 704 can also be used to store one or more running programs and/or modules for performing the data processing methods described above.

The processor 703 may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more An integrated circuit that controls the execution of the program of the present application. The processor 703 can implement the methods provided by the embodiment of FIG. 2 or FIG. 8 above by running or executing software programs and/or modules stored in the memory 704, as well as invoking data stored in the memory 704.

The communication bus 705 can include a path for transferring information between the processor 703 and the memory 704.

In an exemplary embodiment, there is also provided a non-transitory computer readable storage medium comprising instructions, for example, a memory comprising instructions executable by a processor of a first storage device to perform the data storage method described above. For example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device.

FIG. 21 is a schematic structural diagram of a client, which mainly includes a transmitter 801, a receiver 802, a memory 804, a processor 803, and a communication bus 805, according to an exemplary embodiment. It will be understood by those skilled in the art that the structure of the client 800 shown in FIG. 21 does not constitute a limitation on the intelligent lighting device, may include more or less components than those illustrated, or combine some components, or different. The component arrangement is not limited in this embodiment of the present application.

The transmitter 801 can be used to send data, such as frame data, to the subsequent processing module.

The memory 804 can be used to store parsed valid data, and the memory 804 can also be used to store one or more running programs and/or modules for performing the data processing methods described above.

The processor 803 can be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more An integrated circuit that controls the execution of the program of the present application. The processor 803 can implement the method provided by the embodiment of FIG. 2 or FIG. 8 above by running or executing software programs and/or modules stored in the memory 804, as well as invoking data stored in the memory 804.

The communication bus 805 can include a path for transferring information between the processor 803 and the memory 804.

In an exemplary embodiment, there is also provided a non-transitory computer readable storage medium comprising instructions, for example, a memory comprising instructions executable by a processor of a client to perform the data storage method described above. For example, the non-transitory computer readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device.

A person skilled in the art may understand that all or part of the steps of implementing the above embodiments may be completed by hardware, or may be instructed by a program to execute related hardware, and the program may be stored in a computer readable storage medium. The storage medium mentioned may be a read only memory, a magnetic disk or an optical disk or the like.

The above are only the preferred embodiments of the embodiments of the present application, and are not intended to limit the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc., should be made within the spirit and principles of the embodiments of the present application. It is included in the scope of protection of the embodiments of the present application.

Claims (30)

A data storage method is applied to a first storage device of a storage system, where the first storage device includes a plurality of solid state disks, and the method includes: Receiving, by the first storage device, data sent by the client, and performing virtualized strip processing on the data to obtain a plurality of stripe data, where the data sent by the client is the original data written in the client ; The first storage device performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, The verification data is used to verify the plurality of stripe data; The first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives, respectively. The method of claim 1 wherein said first storage device further comprises a plurality of disks, said first storage device said said plurality of stripe data and said said EC strip data After the verification data is separately cached in the plurality of solid state drives, the method further includes: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to data attribute information of the plurality of stripe data to obtain replica data; The first storage device stores the copy data into a plurality of disks included in the first storage device. The method of claim 2, after the storing, by the first storage device, the copy data to the plurality of disks included in the first storage device, further comprising: The first storage device deletes the plurality of stripe data and the check data buffered in the plurality of solid state drives. The method according to claim 1, wherein the first storage device performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC strip data, including: The first storage device generates a vector based on the plurality of stripe data, and multiplies the generated vector by a preset generation matrix to obtain an encoding vector; The first storage device determines data included in the encoding vector as the EC stripe data. The method of claim 4, wherein the method further comprises: When a data loss situation occurs in the plurality of solid state disks, the first storage device determines a target row from the preset generation matrix, and the preset data in the target row is composed of the plurality of stripe data The vector is multiplied to get the missing data; The first storage device deletes the target row in the preset generation matrix to obtain a target generation matrix. The first storage device multiplies a vector composed of EC strip data remaining after data loss with an inverse matrix of the target generation matrix to recover the EC strip data. The method of claim 1, wherein the storage system further comprises a second storage device, and the second storage device comprises a plurality of disks, the first storage device to the EC strip data After the plurality of stripe data and the check data are respectively cached in the plurality of solid state drives, the method further includes: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to data attribute information of the plurality of stripe data to obtain replica data; The first storage device sends the replica data to the second storage device, to store the replica data by the second storage device into a plurality of disks included in the second storage device. A data storage method is applied to a client of a storage system, the storage system further includes a first storage device, and the first storage device includes a plurality of solid state disks, wherein the method includes: The client performs virtualized strip processing on the data to obtain a plurality of stripe data, where the data is original data written in the client; The client performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, the school The test data is used to verify the plurality of stripe data; Transmitting, by the client, the EC strip data to the first storage device, so that the first storage device uses the plurality of stripe data and the check data included in the EC strip data Cache to the plurality of solid state drives respectively. The method according to claim 7, wherein the client performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC strip data, including: The client generates a vector based on the plurality of stripe data, and multiplies the generated vector by a preset generation matrix to obtain a coding vector; The client determines data included in the coding vector as the EC strip data. A data storage method is applied to a first storage device of a storage system, the storage system further includes a client, and the first storage device includes a plurality of solid state disks, wherein the method includes: Receiving, by the first storage device, the EC strip data sent by the client, where the EC strip data includes the plurality of stripe data and check data, where the check data is used for the multiple strips Carrying data for verification, the plurality of stripe data being obtained by the client performing virtualized strip processing on the data, where the data is original data written in the client; The first storage device caches the plurality of stripe data and the check data included in the EC strip data to the plurality of solid state drives, respectively. The method of claim 9, wherein the first storage device further comprises a plurality of disks, the first storage device to the plurality of stripe data included in the EC strip data and the After the verification data is separately cached in the plurality of solid state drives, the method further includes: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to data attribute information of the plurality of stripe data to obtain replica data; The first storage device stores the copy data into a plurality of disks included in the first storage device. The method of claim 10, after the storing, by the first storage device, the copy data to the plurality of disks included in the first storage device, further comprising: The first storage device deletes the plurality of stripe data and the check data buffered in the plurality of solid state drives. The method of claim 9 wherein the method further comprises: When a data loss situation occurs in the plurality of solid state disks, the first storage device determines a target row from the preset generation matrix, and the preset data in the target row is composed of the plurality of stripe data The vector is multiplied to get the missing data; The first storage device deletes the target row in the preset generation matrix to obtain a target generation matrix. The first storage device multiplies a vector composed of EC strip data remaining after data loss with an inverse matrix of the target generation matrix to recover the EC strip data. The method of claim 9, wherein the storage system further comprises a second storage device, and the second storage device comprises a plurality of disks, the first storage device to the EC strip data After the plurality of stripe data and the check data are respectively cached in the plurality of solid state drives, the method further includes: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to data attribute information of the plurality of stripe data to obtain replica data; The first storage device sends the replica data to the second storage device, to store the replica data by the second storage device into a plurality of disks included in the second storage device. A first storage device is deployed in a storage system, where the first storage device includes a plurality of solid state disks, and the first storage device includes: One or more processors; and Memory The memory stores one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including instructions for: Receiving data sent by the client, and performing virtualized strip processing on the data to obtain a plurality of stripe data, where the data sent by the client is the original data written in the client; Encoding the plurality of stripe data by using a codec EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, and the check data is used for Performing verification on the plurality of stripe data; And storing the plurality of stripe data and the check data included in the EC strip data into the plurality of solid state drives respectively. The first storage device of claim 14, wherein the first storage device further comprises a plurality of disks, the one or more programs further comprising instructions for: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the plurality of stripe data included in the EC strip data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data; The copy data is stored in a plurality of disks included in the first storage device. The first storage device of claim 15 wherein said one or more programs further comprise instructions for: And deleting the plurality of stripe data and the check data buffered in the plurality of solid state drives. The first storage device of claim 14, wherein the one or more programs further comprise instructions for: Generating a vector based on the plurality of stripe data, and multiplying the generated vector by a preset generation matrix to obtain a coding vector; The data included in the coding vector is determined as the EC strip data. The first storage device of claim 17 wherein said one or more programs further comprise instructions for: Determining a target row from the preset generation matrix when a data loss condition occurs in the plurality of solid state disks, wherein the preset data in the target row is multiplied by a vector composed of the plurality of stripe data Get lost data; Deleting the target row in the preset generation matrix to obtain a target generation matrix; A vector consisting of the remaining EC strip data after data loss is multiplied by an inverse matrix of the target generation matrix to recover the EC strip data. The first storage device according to claim 14, wherein the storage system further comprises a second storage device, and the second storage device comprises a plurality of disks, and the one or more programs further comprise Instructions for doing the following: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the plurality of stripe data included in the EC strip data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data; Transmitting the copy data to the second storage device to store the copy data by the second storage device to a plurality of disks included in the second storage device. A client is deployed in a storage system, and the storage system further includes a first storage device, where the first storage device includes a plurality of solid state disks, wherein the client includes: One or more processors; and Memory The memory stores one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including instructions for: Performing virtualized stripe processing on the data to obtain a plurality of stripe data, the data being original data written in the client; Encoding the plurality of stripe data by using a codec EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, and the check data is used for Performing verification on the plurality of stripe data; Transmitting the EC stripe data to the first storage device, so that the first storage device caches the plurality of stripe data and the check data included in the EC stripe data respectively Said in multiple solid state drives. The client of claim 20 wherein said one or more programs further comprise instructions for: Generating a vector based on the plurality of stripe data, and multiplying the generated vector by a preset generation matrix to obtain an encoding vector; The data included in the coding vector is determined as the EC strip data. A first storage device is deployed in the storage system, the storage system further includes a client, and the first storage device includes a plurality of solid state disks, wherein the first storage device includes: One or more processors; and Memory The memory stores one or more programs, the one or more programs being configured to be executed by the one or more processors, the one or more programs including instructions for: Receiving EC strip data sent by the client, the EC strip data includes the plurality of stripe data and check data, and the check data is used to verify the plurality of stripe data, The plurality of stripe data is obtained by the client performing virtualized strip processing on the data, where the data is original data written in the client; And storing the plurality of stripe data and the check data included in the EC strip data into the plurality of solid state drives respectively. The first storage device of claim 22, wherein the first storage device further comprises a plurality of disks, the one or more programs further comprising instructions for: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the plurality of stripe data included in the EC strip data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data; The copy data is stored in a plurality of disks included in the first storage device. The first storage device of claim 22, wherein the storage system further comprises a second storage device, and the second storage device comprises a plurality of disks, the one or more programs further comprising Instructions for doing the following: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the plurality of stripe data included in the EC strip data is aggregated into a copy according to the data attribute information of the plurality of stripe data to obtain replica data; Transmitting the copy data to the second storage device to store the copy data by the second storage device to a plurality of disks included in the second storage device. A storage system includes a client and a first storage device, and the first storage device includes a plurality of solid state disks, wherein the method includes: The client performs virtualized strip processing on the data to obtain a plurality of stripe data, where the data is original data written in the client; The client performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC stripe data, where the EC stripe data includes the plurality of stripe data and check data, the school The test data is used to verify the plurality of stripe data; Sending, by the client, the EC strip data to the first storage device; Receiving, by the first storage device, the EC strip data sent by the client, and buffering the plurality of strip data and the check data included in the EC strip data to the plurality of solid state drives respectively in. The storage system according to claim 25, wherein the client performs encoding processing on the plurality of stripe data by using an erasure correction code EC algorithm to obtain EC strip data, including: The client generates a vector based on the plurality of stripe data, and multiplies the generated vector by a preset generation matrix to obtain a coding vector; The client determines data included in the coding vector as the EC strip data. The storage system according to claim 25, wherein said first storage device further comprises a plurality of disks, said first storage device said said plurality of stripe data and said said EC strip data After the verification data is separately cached in the plurality of solid state drives, the method further includes: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to data attribute information of the plurality of stripe data to obtain replica data; The first storage device stores the copy data into a plurality of disks included in the first storage device. The storage system according to claim 27, wherein after the storing, by the first storage device, the copy data to the plurality of disks included in the first storage device, the method further includes: The first storage device deletes the plurality of stripe data and the check data buffered in the plurality of solid state drives. The storage system of claim 25, wherein the first storage device is further configured to: Determining a target row from the preset generation matrix when a data loss condition occurs in the plurality of solid state disks, wherein the preset data in the target row is multiplied by a vector composed of the plurality of stripe data Get lost data; Deleting the target row in the preset generation matrix to obtain a target generation matrix; A vector consisting of the remaining EC strip data after data loss is multiplied by an inverse matrix of the target generation matrix to recover the EC strip data. The storage system according to claim 25, wherein said storage system further comprises a second storage device, and said second storage device comprises a plurality of disks, said first storage device said said EC strip After the plurality of stripe data and the check data included in the data are respectively cached in the plurality of solid state drives, the method further includes: If the data capacity of any one of the plurality of SSDs reaches a preset capacity, or if the data of any one of the plurality of SSDs is read less than the preset time period The preset number of times, the first storage device aggregates the plurality of stripe data included in the EC stripe data into a copy according to data attribute information of the plurality of stripe data to obtain replica data; Transmitting, by the first storage device, the replica data to the second storage device; The second storage device receives the copy data sent by the first storage device, and stores the received copy data into a plurality of disks included in the second storage device.

Images